Commit 624686abcf added an assertion that verified that _bt_search
successfully relocated the leaf page undergoing deletion. Page deletion
cannot deal with the case where the descent stack is to the right of the
page, so this seemed critical (deletion can only handle the case where
the descent stack is to the left of the leaf/target page). However, the
assertion went a bit too far.
Since only a buffer pin is held on the leaf page throughout the call to
_bt_search, nothing guarantees that it can't have split during this
small window. And if does actually split, _bt_search may end up
"relocating" a page to the right of the original target leaf page. This
scenario seems extremely unlikely, but it must still be considered.
Remove the assertion, and document how we cope in this scenario.
Synchronize the event names for parallel hash join waits with other
event names, by getting rid of the slashes and dropping "-ing"
suffixes. Rename ClogGroupUpdate to XactGroupUpdate, to match the
new SLRU name. Move the ProcSignalBarrier event to the IPC category;
it doesn't belong under IO.
Also a bit more wordsmithing in the wait event documentation tables.
Discussion: https://postgr.es/m/4505.1589640417@sss.pgh.pa.us
Choose names that fit into the conventions for wait event names
(particularly, that multi-word names are in the style MultiWordName)
and hopefully convey more information to non-hacker users than the
previous names did.
Also rename SerializablePredicateLockListLock to
SerializablePredicateListLock; the old name was long enough to cause
table formatting problems, plus the double occurrence of "Lock" seems
confusing/error-prone.
Also change a couple of particularly opaque LWLock field names.
Discussion: https://postgr.es/m/28683.1589405363@sss.pgh.pa.us
_bt_killitems marks btree items dead when a scan leaves the page where
they live, but it does so with only share lock (to improve concurrency).
This was historicall okay, since killing a dead item has no
consequences. However, with the advent of data checksums and
wal_log_hints, this action incurs a WAL full-page-image record of the
page. Multiple concurrent processes would write the same page several
times, leading to WAL bloat. The probability of this happening can be
reduced by only killing items if they're not already dead, so change the
code to do that.
The problem could eliminated completely by having _bt_killitems upgrade
to exclusive lock upon seeing a killable item, but that would reduce
concurrency so it's considered a cure worse than the disease.
Backpatch all the way back to 9.5, since wal_log_hints was introduced in
9.4.
Author: Masahiko Sawada <masahiko.sawada@2ndquadrant.com>
Discussion: https://postgr.es/m/CA+fd4k6PeRj2CkzapWNrERkja5G0-6D-YQiKfbukJV+qZGFZ_Q@mail.gmail.com
Originally, the names assigned to SLRUs had no purpose other than
being shmem lookup keys, so not a lot of thought went into them.
As of v13, though, we're exposing them in the pg_stat_slru view and
the pg_stat_reset_slru function, so it seems advisable to take a bit
more care. Rename them to names based on the associated on-disk
storage directories (which fortunately we *did* think about, to some
extent; since those are also visible to DBAs, consistency seems like
a good thing). Also rename the associated LWLocks, since those names
are likewise user-exposed now as wait event names.
For the most part I only touched symbols used in the respective modules'
SimpleLruInit() calls, not the names of other related objects. This
renaming could have been taken further, and maybe someday we will do so.
But for now it seems undesirable to change the names of any globally
visible functions or structs, so some inconsistency is unavoidable.
(But I *did* terminate "oldserxid" with prejudice, as I found that
name both unreadable and not descriptive of the SLRU's contents.)
Table 27.12 needs re-alphabetization now, but I'll leave that till
after the other LWLock renamings I have in mind.
Discussion: https://postgr.es/m/28683.1589405363@sss.pgh.pa.us
Includes some manual cleanup of places that pgindent messed up,
most of which weren't per project style anyway.
Notably, it seems some people didn't absorb the style rules of
commit c9d297751, because there were a bunch of new occurrences
of function calls with a newline just after the left paren, all
with faulty expectations about how the rest of the call would get
indented.
There is little point in using the LWLockRegisterTranche mechanism for
built-in tranche names. It wastes cycles, it creates opportunities for
bugs (since failing to register a tranche name is a very hard-to-detect
problem), and the lack of any centralized list of names encourages
sloppy nonconformity in name choices. Moreover, since we have a
centralized list of the tranches anyway in enum BuiltinTrancheIds, we're
certainly not buying any flexibility in return for these disadvantages.
Hence, nuke all the backend-internal LWLockRegisterTranche calls,
and instead provide a const array of the builtin tranche names.
(I have in mind to change a bunch of these names shortly, but this
patch is just about getting them into one place.)
Discussion: https://postgr.es/m/9056.1589419765@sss.pgh.pa.us
The additional pain from level 4 is excessive for the gain.
Also revert all the source annotation changes to their original
wordings, to avoid back-patching pain.
Discussion: https://postgr.es/m/31166.1589378554@sss.pgh.pa.us
Instead of re-identifying which statistics bucket to use for a given
SLRU on every counter increment, do it once during shmem initialization.
This saves a fair number of cycles, and there's no real cost because
we could not have a bucket assignment that varies over time or across
backends anyway.
Also, get rid of the ill-considered decision to let pgstat.c pry
directly into SLRU's shared state; it's cleaner just to have slru.c
pass the stats bucket number.
In consequence of these changes, there's no longer any need to store
an SLRU's LWLock tranche info in shared memory, so get rid of that,
making this a net reduction in shmem consumption. (That partly
reverts fe702a7b3.)
This is basically code review for 28cac71bd, so I also cleaned up
some comments, removed a dangling extern declaration, fixed some
things that should be static and/or const, etc.
Discussion: https://postgr.es/m/3618.1589313035@sss.pgh.pa.us
* Have both physical and logical walsender share a 'xlogreader' state
struct for tracking state. This replaces the existing globals sendSeg
and sendCxt.
* Change WALRead not to receive XLogReaderState->seg and ->segcxt as
separate arguments anymore; just use the ones from 'state'. This is
made possible by the above change.
* have the XLogReader segment_open contract require the callbacks to
install the file descriptor in the state struct themselves instead of
returning it. xlogreader was already ignoring any possible failed
return from the callbacks, relying solely on them never returning.
(This point is not altogether excellent, as it means the callbacks
have to know more of XLogReaderState; but to really improve on that
we would have to pass back error info from the callbacks to
xlogreader. And the complexity would not be saved but instead just
transferred to the callers of WALRead, which would have to learn how
to throw errors from the open_segment callback in addition of, as
currently, from pg_pread.)
* segment_open no longer receives the 'segcxt' as a separate argument,
since it's part of the XLogReaderState argument.
Per comments from Kyotaro Horiguchi.
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/20200511203336.GA9913@alvherre.pgsql
Use it at level 4, a bit more restrictive than the default level, and
tweak our commanding comments to FALLTHROUGH.
(However, leave zic.c alone, since it's external code; to avoid the
warnings that would appear there, change CFLAGS for that file in the
Makefile.)
Author: Julien Rouhaud <rjuju123@gmail.com>
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/20200412081825.qyo5vwwco3fv4gdo@nol
Discussion: https://postgr.es/m/flat/E1fDenm-0000C8-IJ@gemulon.postgresql.org
The existing callers of XLogReadDetermineTimeline() performing recovery
need to check a replay LSN position when determining on which timeline
to read a WAL page. A portion of the comment describing this function
said exactly that, while referring to a routine for fetching a write
LSN, something not available in recovery.
Author: Kyotaro Horiguchi
Discussion: https://postgr.es/m/20200511.101619.2043820539323292957.horikyota.ntt@gmail.com
Restructure the function that locates the root of the to-be-deleted
subtree during nbtree page deletion. Handle the conditions that make
page deletion unsafe in a slightly more uniform way, and acknowledge the
fact that the behavior with incomplete splits on internal pages is
different (as pointed out in the nbtree README as of commit 35bc0ec7).
Also invent new terminology that avoids ambiguity around which pages are
about to be deleted. Consistently use the term "to-be-deleted subtree",
not the ambiguous term "branch".
We were calling the subtree parent page the "top parent page", but that
was quite misleading. The top parent page usually refers to a page
unlinked from its siblings and marked deleted (during the second stage
of page deletion). There was one kind of top parent page that we merely
removed a downlink from, and another kind of top parent page that we
actually marked deleted. Eliminate the ambiguity by inventing a new
term ("subtree parent page") that refers to the former kind of page
only.
Code review for 0dc8ead463, prompted by a bug closed by 91c40548d5.
XLogReader's system for opening and closing segments had gotten too
complicated, with callbacks being passed at both the XLogReaderAllocate
level (read_page) as well as at the WALRead level (segment_open). This
was confusing and hard to follow, so restructure things so that these
callbacks are passed together at XLogReaderAllocate time, and add
another callback to the set (segment_close) to make it a coherent whole.
Also, ensure XLogReaderState is an argument to all the callbacks, so
that they can grab at the ->private data if necessary.
Document the whole arrangement more clearly.
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Discussion: https://postgr.es/m/20200422175754.GA19858@alvherre.pgsql
TimelineHistoryRead and TimelineHistoryWrite wait events are reported
during waiting for a read and write of a timeline history file, respectively.
However, previously, TimelineHistoryRead wait event was not reported
while readTimeLineHistory() was reading a timeline history file. Also
TimelineHistoryWrite was not reported while writeTimeLineHistory() was
writing one line with the details of the timeline split, at the end.
This commit fixes these issues.
Back-patch to v10 where wait events for a timeline history file was added.
Author: Masahiro Ikeda
Reviewed-by: Michael Paquier, Fujii Masao
Discussion: https://postgr.es/m/d11b0c910b63684424e06772eb844ab5@oss.nttdata.com
Factor out code common to _bt_lock_branch_parent() and _bt_pagedel()
into a new utility function. This new function is used to check that
the left sibling of a deletion target page does not have the
INCOMPLETE_SPLIT page flag set. If it is set then deletion is unsafe;
there won't be a usable pivot tuple (with a downlink) in the parent page
that points to the deletion target page. The page deletion algorithm is
not prepared to deal with that. Also restructure an existing, related
utility function that checks if the right sibling of the target page has
the ISHALFDEAD page flag set.
This organization highlights the symmetry between the two cases. The
goal is to make the design of page deletion clearer. Both functions
involve a sibling page with a flag that indicates that there was an
interrupted operation (a page split or a page deletion) that resulted in
a page pointed to by sibling pages, but not pointed to in the parent.
And, both functions indicate if page deletion is unsafe due to the
absence of a particular downlink in the parent page.
In commit 33e05f89c5, we have added the option to display WAL usage
statistics in Explain and auto_explain. The display format used two spaces
between each field which is inconsistent with Buffer usage statistics which
is using one space between each field. Change the format to make WAL usage
statistics consistent with Buffer usage statistics.
This commit also changed the usage of "full page writes" to
"full page images" for WAL usage statistics to make it consistent with
other parts of code and docs.
Author: Julien Rouhaud, Amit Kapila
Reviewed-by: Justin Pryzby, Kyotaro Horiguchi and Amit Kapila
Discussion: https://postgr.es/m/CAB-hujrP8ZfUkvL5OYETipQwA=e3n7oqHFU=4ZLxWS_Cza3kQQ@mail.gmail.com
Remove one of the arguments to btvacuumpage(), and give up on the idea
that it's a recursive function. We now use the term "backtracking" to
refer to the case where an earlier block must be visited to make sure no
tuples that need to be removed were missed.
Advertising btvacuumpage() as a recursive function was unhelpful. In
reality the function always simulates recursion with a loop (it doesn't
actually call itself). This wasn't just necessary as a precaution (per
the comments mentioning tail recursion), though. There is no reliable
natural limit on the number of times we can backtrack.
There are important behavioral difference when "recursing"/backtracking,
mostly related to page deletion. We don't perform page deletion when
backtracking due to the extra complexity. And when we recurse, we're
not performing a physical order scan anymore, so we expect fairly
different conditions to hold for the page. Structuring the code like
this makes it clearer how _bt_pagedel() cooperates with btvacuumpage()
and btvacuumscan() (as established in commit b0229f26 and commit
73a076b0).
Author: Peter Geoghegan
Reviewed-By: Masahiko Sawada
Discussion: https://postgr.es/m/CAH2-WzmRGMDWiLMcb+zagG9652PboNN4Gfcq1Gc_wJL6A716MA@mail.gmail.com
On further reflection, code comments added by commit b0229f26 slightly
misrepresented how we determine the oldest bpto.xact for the index.
btvacuumpage() does not treat the bpto.xact of a page that it put in the
FSM as a candidate to be the oldest deleted page (the delete-marked page
that has the oldest bpto.xact XID among all pages encountered).
The definition of a deleted page for the purposes of the bpto.xact
calculation is different from the definition used by the bulk delete
statistics. The bulk delete statistics don't distinguish between pages
that were deleted by the current VACUUM, pages deleted by a previous
VACUUM operation but not yet recyclable/reusable, and pages that are
reusable (though reusable pages are counted separately).
Backpatch: 11-, just like commit b0229f26.
The logic for determining how many nbtree pages in an index are deleted
pages sometimes undercounted pages. Pages that were deleted by the
current VACUUM operation (as opposed to some previous VACUUM operation
whose deleted pages have yet to be reused) were sometimes overlooked.
The final count is exposed to users through VACUUM VERBOSE's "%u index
pages have been deleted" output.
btvacuumpage() avoided double-counting when _bt_pagedel() deleted more
than one page by assuming that only one page was deleted, and that the
additional deleted pages would get picked up during a future call to
btvacuumpage() by the same VACUUM operation. _bt_pagedel() can
legitimately delete pages that the btvacuumscan() scan will not visit
again, though, so that assumption was slightly faulty.
Fix the accounting by teaching _bt_pagedel() about its caller's
requirements. It now only reports on pages that it knows btvacuumscan()
won't visit again (including the current btvacuumpage() page), so
everything works out in the end.
This bug has been around forever. Only backpatch to v11, though, to
keep _bt_pagedel() is sync on the branches that have today's bugfix
commit b0229f26da. Note that this commit changes the signature of
_bt_pagedel(), just like commit b0229f26da.
Author: Peter Geoghegan
Reviewed-By: Masahiko Sawada
Discussion: https://postgr.es/m/CAH2-WzkrXBcMQWAYUJMFTTvzx_r4q=pYSjDe07JnUXhe+OZnJA@mail.gmail.com
Backpatch: 11-
Commit 857f9c36cd (which taught nbtree VACUUM to skip a scan of the
index from btcleanup in situations where it doesn't seem worth it) made
VACUUM maintain the oldest btpo.xact among all deleted pages for the
index as a whole. It failed to handle all the details surrounding pages
that are deleted by the current VACUUM operation correctly (though pages
deleted by some previous VACUUM operation were processed correctly).
The most immediate problem was that the special area of the page was
examined without a buffer pin at one point. More fundamentally, the
handling failed to account for the full range of _bt_pagedel()
behaviors. For example, _bt_pagedel() sometimes deletes internal pages
in passing, as part of deleting an entire subtree with btvacuumpage()
caller's page as the leaf level page. The original leaf page passed to
_bt_pagedel() might not be the page that it deletes first in cases where
deletion can take place.
It's unclear how disruptive this bug may have been, or what symptoms
users might want to look out for. The issue was spotted during
unrelated code review.
To fix, push down the logic for maintaining the oldest btpo.xact to
_bt_pagedel(). btvacuumpage() is now responsible for pages that were
fully deleted by a previous VACUUM operation, while _bt_pagedel() is now
responsible for pages that were deleted by the current VACUUM operation
(this includes half-dead pages from a previous interrupted VACUUM
operation that become fully deleted in _bt_pagedel()). Note that
_bt_pagedel() should never encounter an existing deleted page.
This commit theoretically breaks the ABI of a stable release by changing
the signature of _bt_pagedel(). However, if any third party extension
is actually affected by this, then it must already be completely broken
(since there are numerous assumptions made in _bt_pagedel() that cannot
be met outside of VACUUM). It seems highly unlikely that such an
extension actually exists, in any case.
Author: Peter Geoghegan
Reviewed-By: Masahiko Sawada
Discussion: https://postgr.es/m/CAH2-WzkrXBcMQWAYUJMFTTvzx_r4q=pYSjDe07JnUXhe+OZnJA@mail.gmail.com
Backpatch: 11-, where the "skip full scan" feature was introduced.
Avoid accessing the leaf page's top parent tuple without a buffer lock
held during the second phase of nbtree page deletion. The old approach
was safe, though only because VACUUM never drops its buffer pin (and
because only VACUUM itself can modify a half-dead page). Even still, it
seems like a good idea to be strict here. Tighten things up by copying
the top parent page's block number to a local variable before releasing
the buffer lock on the leaf page -- not after.
This is a follow-up to commit fa7ff642, which fixed a similar issue in
the first phase of nbtree page deletion.
Update some related comments in passing.
Discussion: https://postgr.es/m/CAH2-WzkLgyN3zBvRZ1pkNJThC=xi_0gpWRUb_45eexLH1+k2_Q@mail.gmail.com
Avoid accessing the deletion target page's special area during nbtree
page deletion at a point where there is no buffer lock held. This issue
was detected by a patch that extends Valgrind's memcheck tool to mark
nbtree pages that are unsafe to access (due to not having a buffer lock
or buffer pin) as NOACCESS.
We do hold a buffer pin at this point, and only access the special area,
so the old approach was safe. Even still, it seems like a good idea to
tighten up the rules in this area. There is no reason to not simply
insist on always holding a buffer lock (not just a pin) when accessing
nbtree pages.
Update some related comments in passing.
Discussion: https://postgr.es/m/CAH2-WzkLgyN3zBvRZ1pkNJThC=xi_0gpWRUb_45eexLH1+k2_Q@mail.gmail.com
78ea8b5 has fixed an issue related to the recycling of WAL segments on
standbys depending on archive_mode. However, it has introduced a
regression with the handling of WAL segments ready to be archived during
crash recovery, causing those files to be recycled without getting
archived.
This commit fixes the regression by tracking in shared memory if a live
cluster is either in crash recovery or archive recovery as the handling
of WAL segments ready to be archived is different in both cases (those
WAL segments should not be removed during crash recovery), and by using
this new shared memory state to decide if a segment can be recycled or
not. Previously, it was not possible to know if a cluster was in crash
recovery or archive recovery as the shared state was able to track only
if recovery was happening or not, leading to the problem.
A set of TAP tests is added to close the gap here, making sure that WAL
segments ready to be archived are correctly handled when a cluster is in
archive or crash recovery with archive_mode set to "on" or "always", for
both standby and primary.
Reported-by: Benoît Lobréau
Author: Jehan-Guillaume de Rorthais
Reviewed-by: Kyotaro Horiguchi, Fujii Masao, Michael Paquier
Discussion: https://postgr.es/m/20200331172229.40ee00dc@firost
Backpatch-through: 9.5
index.c supposed that it could just use a PG_TRY block to clean up the
state associated with an active REINDEX operation. However, that code
doesn't run if we do a FATAL exit --- for example, due to a SIGTERM
shutdown signal --- while the REINDEX is happening. And that state does
get consulted during catalog accesses, which makes it problematic if we
do any catalog accesses during shutdown --- for example, to clean up any
temp tables created in the session.
If this combination of circumstances occurred, we could find ourselves
trying to access already-freed memory. In debug builds that'd fairly
reliably cause an assertion failure. In production we might often
get away with it, but with some bad luck it could cause a core dump.
Another possible bad outcome is an erroneous conclusion that an
index-to-be-accessed is being reindexed; but it looks like that would
be unlikely to have any consequences worse than failing to drop temp
tables right away. (They'd still get dropped by the next session that
uses that temp schema.)
To fix, get rid of the use of PG_TRY here, and instead hook into
the transaction abort mechanisms to clean up reindex state.
Per bug #16378 from Alexander Lakhin. This has been wrong for a
very long time, so back-patch to all supported branches.
Discussion: https://postgr.es/m/16378-7a70ca41b3ec2009@postgresql.org
Commit 0d861bbb, which introduced deduplication to nbtree, added some
logic to take large posting list tuples into account when choosing a
split point. We subtract firstright posting list overhead from the
projected new high key size when calculating leftfree/rightfree values
for an affected candidate split point. Posting list tuples aren't
special to nbtsplitloc.c, but taking them into account like this makes a
huge difference in practice. Posting list tuples are frequently tuple
size outliers.
However, commit 0d861bbb missed a closely related issue: split interval
itself is calculated based on the assumption that tuples on the page
being split are roughly equisized. That assumption was acceptable back
when commit fab25024 taught the logic for choosing a split point about
suffix truncation, but it's pretty questionable now that very large
tuple sizes are common. This oversight led to unbalanced page splits in
low cardinality multi-column indexes when deduplication was used: page
splits that don't give sufficient weight to how unbalanced the split is
when the interval happens to include some large posting list tuples (and
when most other tuples on the page are not so large).
Nail this down by calculating an initial split interval in a way that's
attuned to the actual cost that we want to keep under control (not a
fuzzy proxy for the cost): apply a leftfree + rightfree evenness test to
each candidate split point that actually gets included in the split
interval (for the default strategy). This replaces logic that used a
percentage of all legal split points for the page as the basis of the
initial split interval.
Discussion: https://postgr.es/m/CAH2-WznJt5aT2uUB2Bs+JBLdwe0XTX67+xeLFcaNvCKxO=QBVQ@mail.gmail.com
A comment from the Berkeley days incorrectly claimed that the page
management code cares about the contents of the hole in the center of
the page (at least in the case of the left half of an nbtree page
split). Commit 8fa30f906b added an addendum that stated that the
original comment was "probably obsolete". It's definitely obsolete,
though, so remove the original comment plus the addendum.
Nest the "update metapage as part of insert into root-like page" branch
inside the broader "insert into internal page" branch. This improves
readability.
Clearly it's not okay for nbtree to split a page that is the only page
on its level, and then find that it has to split the parent one level up
in turn. There is simply no code to handle the split_only_page case in
the _bt_insertonpg() "newitem won't fit" branch (only the "newitem fits"
branch handles split_only_page). Add a defensive assertion that will
fail if a split_only_page call to _bt_insertonpg() somehow ends up
splitting the target/parent page.
I (pgeoghegan) believe that we don't need split_only_page handling for
the "newitem won't fit" branch because anybody calling _bt_insertonpg()
like this would have to hold a lock on the same one and only child page.
It seems like a good idea for nbtree's retail insert code to be
absolutely consistent with nbtree's page split code for anything that
naturally requires equivalent handling. Anything that concerns
inserting newitem (which is handled as part of the page split atomic
action when a page split is required) should work in exactly the same
way. With that in mind, make _bt_insertonpg() handle 'cbuf' in a way
that matches _bt_split().
An nbtree split point can be thought of as a point between two adjoining
tuples from an imaginary version of the page being split that includes
the incoming/new item (in addition to the items that really are on the
page). These adjoining tuples are called the lastleft and firstright
tuples.
The variables that represent split points contained a field called
firstright, which is an offset number of the first data item from the
original page that goes on the new right page. The corresponding tuple
from origpage was usually the same thing as the actual firstright tuple,
but not always: the firstright tuple is sometimes the new/incoming item
instead. This situation seems unnecessarily confusing.
Make things clearer by renaming the origpage offset returned by
_bt_findsplitloc() to "firstrightoff". We now have a firstright tuple
and a firstrightoff offset number which are comparable to the
newitem/lastleft tuples and the newitemoff/lastleftoff offset numbers
respectively. Also make sure that we are consistent about how we
describe nbtree page split point state.
Push the responsibility for dealing with pg_upgrade'd !heapkeyspace
indexes down to lower level code, relieving _bt_split() from dealing
with it directly. This means that we always have a palloc'd left page
high key on the leaf level, no matter what. This enables simplifying
some of the code (and code comments) within _bt_split().
Finally, restructure the page split code to make it clearer why suffix
truncation (which only takes place during leaf page splits) is
completely different to the first data item truncation that takes place
during internal page splits. Tuples are marked as having fewer
attributes stored in both cases, and the firstright tuple is truncated
in both cases, so it's easy to imagine somebody missing the distinction.
Currently, we don't account for buffer usage incurred by parallel workers
for parallel create index. This commit allows each worker to record the
buffer usage stats and leader backend to accumulate that stats at the
end of the operation. This will allow pg_stat_statements to display
correct buffer usage stats for (parallel) create index command.
Reported-by: Julien Rouhaud
Author: Sawada Masahiko
Reviewed-by: Dilip Kumar, Julien Rouhaud and Amit Kapila
Backpatch-through: 11, where this was introduced
Discussion: https://postgr.es/m/20200328151721.GB12854@nol
GetWalRcvWriteRecPtr() previously reported the latest *flushed*
location. Adopt the conventional terminology used elsewhere in the tree
by renaming it to GetWalRcvFlushRecPtr(), and likewise for some related
variables that used the term "received".
Add a new definition of GetWalRcvWriteRecPtr(), which returns the latest
*written* value. This will allow later patches to use the value for
non-data-integrity purposes, without having to wait for the flush
pointer to advance.
Reviewed-by: Alvaro Herrera <alvherre@2ndquadrant.com>
Reviewed-by: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/CA%2BhUKGJ4VJN8ttxScUFM8dOKX0BrBiboo5uz1cq%3DAovOddfHpA%40mail.gmail.com
0f5ca02f53 introduces 3 new keywords. It appears to be too much for relatively
small feature. Given now we past feature freeze, it's already late for
discussion of the new syntax. So, revert.
Discussion: https://postgr.es/m/28209.1586294824%40sss.pgh.pa.us
2nd pass of modifying various places which obtain the next power
of 2 of a number and make them use the new functions added in
f0705bb62.
In passing, also modify num_combinations(). This can be implemented
using simple bitshifting rather than looping.
Reviewed-by: John Naylor
Discussion: https://postgr.es/m/20200114173553.GE32763%40fetter.org
First pass of modifying various places that obtain the next power of 2 of
a number and make them use the new functions added in pg_bitutils.h
instead.
This also removes the _hash_log2() function. There are no longer any
callers in core. Other users can swap their _hash_log2(n) call to make use
of pg_ceil_log2_32(n).
Author: David Fetter, with some minor adjustments by me
Reviewed-by: John Naylor, Jesse Zhang
Discussion: https://postgr.es/m/20200114173553.GE32763%40fetter.org
The goal of separating hotly accessed per-backend data from PGPROC
into PGXACT is to make accesses fast (GetSnapshotData() in
particular). But delayChkpt is not actually accessed frequently; only
when starting a checkpoint. As it is frequently modified (multiple
times in the course of a single transaction), storing it in the same
cacheline as hotly accessed data unnecessarily dirties a contended
cacheline.
Therefore move delayChkpt to PGPROC.
This is part of a larger series of patches intending to improve
GetSnapshotData() scalability. It is committed and pushed separately,
as it is independently beneficial (small but measurable win, limited
by the other frequent modifications of PGXACT).
Author: Andres Freund
Reviewed-By: Robert Haas, Thomas Munro, David Rowley
Discussion: https://postgr.es/m/20200301083601.ews6hz5dduc3w2se@alap3.anarazel.de
SLRU page hits were tracked only in SimpleLruReadPage, but that's not
enough because we may hit the page in SimpleLruReadPage_ReadOnly in
which case we don't call SimpleLruReadPage at all.
Reported-by: Kuntal Ghosh
Discussion: https://postgr.es/m/20200119143707.gyinppnigokesjok@development
Before the fix every 2PC commit/abort leaked a file descriptor. As the
files are opened using BasicOpenFile(), that quickly leads to the
backend running out of file descriptors.
Once enough 2PC abort/commit have caused enough FDs to leak, any IO
in the backend will fail with "Too many open files", as
BasicOpenFilePerm() will have triggered all open files known to fd.c
to be closed.
The leak causing the problem at hand is a consequence of 0dc8ead46,
but is only exascerbated by it. Previously most XLogPageReadCB
callbacks used static variables to cache one open file, but after the
commit the cache is private to each XLogReader instance. There never
was infrastructure to close FDs at the time of XLogReaderFree, but the
way XLogReader was used limited the leak to one FD.
This commit just closes the during XLogReaderFree() if the FD is
stored in XLogReaderState.seg.ws_segno. This may not be the way to
solve this medium/long term, but at least unbreaks 2PC.
Discussion: https://postgr.es/m/20200406025651.fpzdb5yyb7qyhqko@alap3.anarazel.de
Since heap TID is supposed to be just another key attribute to the
implementation, it doesn't make much sense to have separate
BTreeTupleSetNAtts() and BTreeTupleSetAltHeapTID() functions. Merge the
two functions together. This slightly simplifies _bt_truncate().
Replication slots are useful to retain data that may be needed by a
replication system. But experience has shown that allowing them to
retain excessive data can lead to the primary failing because of running
out of space. This new feature allows the user to configure a maximum
amount of space to be reserved using the new option
max_slot_wal_keep_size. Slots that overrun that space are invalidated
at checkpoint time, enabling the storage to be released.
Author: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>
Reviewed-by: Masahiko Sawada <sawada.mshk@gmail.com>
Reviewed-by: Jehan-Guillaume de Rorthais <jgdr@dalibo.com>
Reviewed-by: Álvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/20170228.122736.123383594.horiguchi.kyotaro@lab.ntt.co.jp
This commit adds following optional clause to BEGIN and START TRANSACTION
commands.
WAIT FOR LSN lsn [ TIMEOUT timeout ]
New clause pospones transaction start till given lsn is applied on standby.
This clause allows user be sure, that changes previously made on primary would
be visible on standby.
New shared memory struct is used to track awaited lsn per backend. Recovery
process wakes up backend once required lsn is applied.
Author: Ivan Kartyshov, Anna Akenteva
Reviewed-by: Craig Ringer, Thomas Munro, Robert Haas, Kyotaro Horiguchi
Reviewed-by: Masahiko Sawada, Ants Aasma, Dmitry Ivanov, Simon Riggs
Reviewed-by: Amit Kapila, Alexander Korotkov
Discussion: https://postgr.es/m/0240c26c-9f84-30ea-fca9-93ab2df5f305%40postgrespro.ru
Previously when there were multiple timelines listed in the history file
of the recovery target timeline, archive recovery searched all of them,
starting from the newest timeline to the oldest one, to find the segment
to read. That is, archive recovery had to continuously fail scanning
the segment until it reached the timeline that the segment belonged to.
These scans for non-existent segment could be harmful on the recovery
performance especially when archival area was located on the remote
storage and each scan could take a long time.
To address the issue, this commit changes archive recovery so that
it skips scanning the timeline that the segment to read doesn't belong to.
Author: Kyotaro Horiguchi, tweaked a bit by Fujii Masao
Reviewed-by: David Steele, Pavel Suderevsky, Grigory Smolkin
Discussion: https://postgr.es/m/16159-f5a34a3a04dc67e0@postgresql.org
Discussion: https://postgr.es/m/20200129.120222.1476610231001551715.horikyota.ntt@gmail.com
Similar to xid, but 64 bits wide. This new type is suitable for use in
various system views and administration functions.
Reviewed-by: Fujii Masao <masao.fujii@oss.nttdata.com>
Reviewed-by: Takao Fujii <btfujiitkp@oss.nttdata.com>
Reviewed-by: Yoshikazu Imai <imai.yoshikazu@fujitsu.com>
Reviewed-by: Mark Dilger <mark.dilger@enterprisedb.com>
Discussion: https://postgr.es/m/20190725000636.666m5mad25wfbrri%40alap3.anarazel.de
An assertion added by commit 0d861bbb checked that _bt_killitems() only
processes a BTScanPosItem whose heap TID is contained in a posting list
tuple when its page offset number still matches what is on the page
(i.e. when it matches the posting list tuple's current offset number).
This was only correct in the common case where the page can't have
changed since we first read it. It was not correct in cases where we
don't drop the buffer pin (and don't need to verify the page hasn't
changed using its LSN). The latter category includes scans involving
unlogged tables, and scans that use a non-MVCC snapshot, per the logic
originally introduced by commit 2ed5b87f.
The assertion still seems helpful. Fix it by taking cases where the
page may have been concurrently modified into account.
Reported-By: Anastasia Lubennikova, Alexander Lakhin
Discussion: https://postgr.es/m/c4e38e9a-0f9c-8e53-e639-adf343f94472@postgrespro.ru
There's a very low risk that RecentGlobalXmin could be far enough in
the past to be older than relfrozenxid, or even wrapped
around. Luckily the consequences of that having happened wouldn't be
too bad - the page wouldn't be pruned for a while.
Avoid that risk by using TransactionXmin instead. As that's announced
via MyPgXact->xmin, it is protected against wrapping around (see code
comments for details around relfrozenxid).
Author: Andres Freund
Discussion: https://postgr.es/m/20200328213023.s4eyijhdosuc4vcj@alap3.anarazel.de
Backpatch: 9.5-
Until now, only selected bulk operations (e.g. COPY) did this. If a
given relfilenode received both a WAL-skipping COPY and a WAL-logged
operation (e.g. INSERT), recovery could lose tuples from the COPY. See
src/backend/access/transam/README section "Skipping WAL for New
RelFileNode" for the new coding rules. Maintainers of table access
methods should examine that section.
To maintain data durability, just before commit, we choose between an
fsync of the relfilenode and copying its contents to WAL. A new GUC,
wal_skip_threshold, guides that choice. If this change slows a workload
that creates small, permanent relfilenodes under wal_level=minimal, try
adjusting wal_skip_threshold. Users setting a timeout on COMMIT may
need to adjust that timeout, and log_min_duration_statement analysis
will reflect time consumption moving to COMMIT from commands like COPY.
Internally, this requires a reliable determination of whether
RollbackAndReleaseCurrentSubTransaction() would unlink a relation's
current relfilenode. Introduce rd_firstRelfilenodeSubid. Amend the
specification of rd_createSubid such that the field is zero when a new
rel has an old rd_node. Make relcache.c retain entries for certain
dropped relations until end of transaction.
Bump XLOG_PAGE_MAGIC, since this introduces XLOG_GIST_ASSIGN_LSN.
Future servers accept older WAL, so this bump is discretionary.
Kyotaro Horiguchi, reviewed (in earlier, similar versions) by Robert
Haas. Heikki Linnakangas and Michael Paquier implemented earlier
designs that materially clarified the problem. Reviewed, in earlier
designs, by Andrew Dunstan, Andres Freund, Alvaro Herrera, Tom Lane,
Fujii Masao, and Simon Riggs. Reported by Martijn van Oosterhout.
Discussion: https://postgr.es/m/20150702220524.GA9392@svana.org
This allows gathering the WAL generation statistics for each statement
execution. The three statistics that we collect are the number of WAL
records, the number of full page writes and the amount of WAL bytes
generated.
This helps the users who have write-intensive workload to see the impact
of I/O due to WAL. This further enables us to see approximately what
percentage of overall WAL is due to full page writes.
In the future, we can extend this functionality to allow us to compute the
the exact amount of WAL data due to full page writes.
This patch in itself is just an infrastructure to compute WAL usage data.
The upcoming patches will expose this data via explain, auto_explain,
pg_stat_statements and verbose (auto)vacuum output.
Author: Kirill Bychik, Julien Rouhaud
Reviewed-by: Dilip Kumar, Fujii Masao and Amit Kapila
Discussion: https://postgr.es/m/CAB-hujrP8ZfUkvL5OYETipQwA=e3n7oqHFU=4ZLxWS_Cza3kQQ@mail.gmail.com
A manifest is a JSON document which includes (1) the file name, size,
last modification time, and an optional checksum for each file backed
up, (2) timelines and LSNs for whatever WAL will need to be replayed
to make the backup consistent, and (3) a checksum for the manifest
itself. By default, we use CRC-32C when checksumming data files,
because we are trying to detect corruption and user error, not foil an
adversary. However, pg_basebackup and the server-side BASE_BACKUP
command now have options to select a different algorithm, so users
wanting a cryptographic hash function can select SHA-224, SHA-256,
SHA-384, or SHA-512. Users not wanting file checksums at all can
disable them, or disable generating of the backup manifest altogether.
Using a cryptographic hash function in place of CRC-32C consumes
significantly more CPU cycles, which may slow down backups in some
cases.
A new tool called pg_validatebackup can validate a backup against the
manifest. If no checksums are present, it can still check that the
right files exist and that they have the expected sizes. If checksums
are present, it can also verify that each file has the expected
checksum. Additionally, it calls pg_waldump to verify that the
expected WAL files are present and parseable. Only plain format
backups can be validated directly, but tar format backups can be
validated after extracting them.
Robert Haas, with help, ideas, review, and testing from David Steele,
Stephen Frost, Andrew Dunstan, Rushabh Lathia, Suraj Kharage, Tushar
Ahuja, Rajkumar Raghuwanshi, Mark Dilger, Davinder Singh, Jeevan
Chalke, Amit Kapila, Andres Freund, and Noah Misch.
Discussion: http://postgr.es/m/CA+TgmoZV8dw1H2bzZ9xkKwdrk8+XYa+DC9H=F7heO2zna5T6qg@mail.gmail.com
entryGetItem()'s three code paths each contained bugs associated
with filtering the entries for gin_fuzzy_search_limit.
The posting-tree path failed to advance "advancePast" after having
decided to filter an item. If we ran out of items on the current
page and needed to advance to the next, what would actually happen
is that entryLoadMoreItems() would re-load the same page. Eventually,
the random dropItem() test would accept one of the same items it'd
previously rejected, and we'd move on --- but it could take awhile
with small gin_fuzzy_search_limit. To add insult to injury, this
case would inevitably cause entryLoadMoreItems() to decide it needed
to re-descend from the root, making things even slower.
The posting-list path failed to implement gin_fuzzy_search_limit
filtering at all, so that all entries in the posting list would
be returned.
The bitmap-result path used a "gotitem" variable that it failed to
update in the one place where it'd actually make a difference, ie
at the one "continue" statement. I think this was unreachable in
practice, because if we'd looped around then it shouldn't be the
case that the entries on the new page are before advancePast.
Still, the "gotitem" variable was contributing nothing to either
clarity or correctness, so get rid of it.
Refactor all three loops so that the termination conditions are
more alike and less unreadable.
The code coverage report showed that we had no coverage at all for
the re-descend-from-root code path in entryLoadMoreItems(), which
seems like a very bad thing, so add a test case that exercises it.
We also had exactly no coverage for gin_fuzzy_search_limit, so add a
simplistic test case that at least hits those code paths a little bit.
Back-patch to all supported branches.
Adé Heyward and Tom Lane
Discussion: https://postgr.es/m/CAEknJCdS-dE1Heddptm7ay2xTbSeADbkaQ8bU2AXRCVC2LdtKQ@mail.gmail.com
Commit 40d964ec99 allowed vacuum command to process indexes in parallel but
forgot to accumulate the buffer usage stats of parallel workers. This
allows leader backend to accumulate buffer usage stats of all the parallel
workers.
Reported-by: Julien Rouhaud
Author: Sawada Masahiko
Reviewed-by: Dilip Kumar, Amit Kapila and Julien Rouhaud
Discussion: https://postgr.es/m/20200328151721.GB12854@nol
There's a number of SLRU caches used to access important data like clog,
commit timestamps, multixact, asynchronous notifications, etc. Until now
we had no easy way to monitor these shared caches, compute hit ratios,
number of reads/writes etc.
This commit extends the statistics collector to track this information
for a predefined list of SLRUs, and also introduces a new system view
pg_stat_slru displaying the data.
The list of built-in SLRUs is fixed, but additional SLRUs may be defined
in extensions. Unfortunately, there's no suitable registry of SLRUs, so
this patch simply defines a fixed list of SLRUs with entries for the
built-in ones and one entry for all additional SLRUs. Extensions adding
their own SLRU are fairly rare, so this seems acceptable.
This patch only allows monitoring of SLRUs, not tuning. The SLRU sizes
are still fixed (hard-coded in the code) and it's not entirely clear
which of the SLRUs might need a GUC to tune size. In a way, allowing us
to determine that is one of the goals of this patch.
Bump catversion as the patch introduces new functions and system view.
Author: Tomas Vondra
Reviewed-by: Alvaro Herrera
Discussion: https://www.postgresql.org/message-id/flat/20200119143707.gyinppnigokesjok@development
When recovery target is reached and recovery is paused because of
recovery_target_action=pause, executing pg_wal_replay_resume() causes
the standby to promote, i.e., the recovery to end. So, in this case,
the previous message "Execute pg_wal_replay_resume() to continue"
logged was confusing because pg_wal_replay_resume() doesn't cause
the recovery to continue.
This commit improves the message logged when recovery is paused,
and the proper message is output based on what (pg_wal_replay_pause
or recovery_target_action) causes recovery to be paused.
Author: Sergei Kornilov, revised by Fujii Masao
Reviewed-by: Robert Haas
Discussion: https://postgr.es/m/19168211580382043@myt5-b646bde4b8f3.qloud-c.yandex.net
The definitions of the routines defined in xlogarchive.c have been part
of xlog_internal.h which is included by several frontend tools, but all
those routines are only called by the backend. More cleanup could be
done within xlog_internal.h, but that's already a nice cut.
This will help a follow-up patch for pg_rewind where handling of
restore_command is added for frontends.
Author: Alexey Kondratov, Michael Paquier
Reviewed-by: Álvaro Herrera, Alexander Korotkov
Discussion: https://postgr.es/m/a3acff50-5a0d-9a2c-b3b2-ee36168955c1@postgrespro.ru
vacuum code.
After commit b61d161c14, during vacuum, we cache the information of
relation name and relation namespace in local structure LVRelStats so that
we can use it in an error callback function. We can use the cached
information to avoid the calls to RelationGetRelationName(),
RelationGetNamespace() and get_namespace_name(). This is mainly for the
consistent in vacuum code path but it will avoid the extra syscache lookup
we do in get_namespace_name().
Author: Justin Pryzby
Reviewed-by: Amit Kapila
Discussion: https://www.postgresql.org/message-id/20191120210600.GC30362@telsasoft.com
Commit 7c2dbc69 reorganized _bt_truncate() in a way that enables a
further simplification that I (pgeoghegan) missed: Since we mark the
tuple that is returned to the caller as a pivot tuple before the point
where its heap TID is set as of 7c2dbc69, it is possible to use the high
level BTreeTupleGetHeapTID() inline function to get an item pointer. Do
it that way now. This approach is clearer and more maintainable.
This reverts commit 2aa6e33, that added a fast path to skip
anti-wraparound and non-aggressive autovacuum jobs (these have no sense
as anti-wraparound implies aggressive). With a cluster using a high
amount of relations with a portion of them being heavily updated, this
could cause autovacuum to lock down, with autovacuum workers attempting
repeatedly those jobs on the same relations for the same database, that
just kept being skipped. This lock down can be solved with a manual
VACUUM FREEZE.
Justin King has reported one environment where the issue happened, and
Julien Rouhaud and I have been able to reproduce it in a second
environment. With a very aggressive autovacuum_freeze_max_age,
triggering those jobs with pgbench is a matter of minutes, and hitting
the lock down is a lot harder (my local tests failed to do that).
Note that anti-wraparound and non-aggressive jobs can only be triggered
on a subset of shared catalogs:
- pg_auth_members
- pg_authid
- pg_database
- pg_replication_origin
- pg_shseclabel
- pg_subscription
- pg_tablespace
While the lock down was possible down to v12, the root cause of those
jobs is a much older issue, which needs more analysis.
Bonus thanks to Andres Freund for the discussion.
Reported-by: Justin King
Discussion: https://postgr.es/m/CAE39h22zPLrkH17GrkDgAYL3kbjvySYD1io+rtnAUFnaJJVS4g@mail.gmail.com
Backpatch-through: 12
Simplify _bt_truncate(), the routine that generates truncated leaf page
high keys. Remove a micro-optimization that avoided a second palloc0()
call (this was used when a heap TID was needed in the final pivot tuple,
though only when the index happened to not be an INCLUDE index).
Removing this dubious micro-optimization allows _bt_truncate() to use
the index_truncate_tuple() indextuple.c utility routine in all cases.
This was already the common case.
This commit is a HEAD-only follow up to bugfix commit 4b42a899.
The recheck isn't needed anymore, as RelationGetBufferForTuple() now
extends the relation with RBM_ZERO_AND_LOCK. Previously we needed to
handle the fact that relation extension extended the relation and then
separately acquired a lock on the page - while expecting that the page
is empty.
Reported-By: Ranier Vilela
Discussion: https://postgr.es/m/CAEudQArA_=J0D5T258xsCY6Xtf6wiH4b=QDPDgVS+WZUN10WDw@mail.gmail.com
Early versions of opclass options patch uses zero support procedure as opclass
options procedure. This commit removes rudiments of it, which were committed
in 911e702077. Also, it implements correct handling of amoptsprocnum == 0.
INCLUDE indexes failed to have their non-key attributes physically
truncated away in certain rare cases. This led to physically larger
pivot tuples that contained useless non-key attribute values. The
impact on users should be negligible, but this is still clearly a
regression (Postgres 11 supports INCLUDE indexes, and yet was not
affected).
The bug appeared in commit dd299df8, which introduced "true" suffix
truncation of key attributes.
Discussion: https://postgr.es/m/CAH2-Wz=E8pkV9ivRSFHtv812H5ckf8s1-yhx61_WrJbKccGcrQ@mail.gmail.com
Backpatch: 12-, where "true" suffix truncation was introduced.
PostgreSQL provides set of template index access methods, where opclasses have
much freedom in the semantics of indexing. These index AMs are GiST, GIN,
SP-GiST and BRIN. There opclasses define representation of keys, operations on
them and supported search strategies. So, it's natural that opclasses may be
faced some tradeoffs, which require user-side decision. This commit implements
opclass parameters allowing users to set some values, which tell opclass how to
index the particular dataset.
This commit doesn't introduce new storage in system catalog. Instead it uses
pg_attribute.attoptions, which is used for table column storage options but
unused for index attributes.
In order to evade changing signature of each opclass support function, we
implement unified way to pass options to opclass support functions. Options
are set to fn_expr as the constant bytea expression. It's possible due to the
fact that opclass support functions are executed outside of expressions, so
fn_expr is unused for them.
This commit comes with some examples of opclass options usage. We parametrize
signature length in GiST. That applies to multiple opclasses: tsvector_ops,
gist__intbig_ops, gist_ltree_ops, gist__ltree_ops, gist_trgm_ops and
gist_hstore_ops. Also we parametrize maximum number of integer ranges for
gist__int_ops. However, the main future usage of this feature is expected
to be json, where users would be able to specify which way to index particular
json parts.
Catversion is bumped.
Discussion: https://postgr.es/m/d22c3a18-31c7-1879-fc11-4c1ce2f5e5af%40postgrespro.ru
Author: Nikita Glukhov, revised by me
Reviwed-by: Nikolay Shaplov, Robert Haas, Tom Lane, Tomas Vondra, Alvaro Herrera
When certain parameters are changed on a physical replication primary,
this is communicated to standbys using the XLOG_PARAMETER_CHANGE WAL
record. The standby then checks whether its own settings are at least
as big as the ones on the primary. If not, the standby shuts down
with a fatal error.
The correspondence of settings between primary and standby is required
because those settings influence certain shared memory sizings that
are required for processing WAL records that the primary might send.
For example, if the primary sends a prepared transaction, the standby
must have had max_prepared_transaction set appropriately or it won't
be able to process those WAL records.
However, fatally shutting down the standby immediately upon receipt of
the parameter change record might be a bit of an overreaction. The
resources related to those settings are not required immediately at
that point, and might never be required if the activity on the primary
does not exhaust all those resources. If we just let the standby roll
on with recovery, it will eventually produce an appropriate error when
those resources are used.
So this patch relaxes this a bit. Upon receipt of
XLOG_PARAMETER_CHANGE, we still check the settings but only issue a
warning and set a global flag if there is a problem. Then when we
actually hit the resource issue and the flag was set, we issue another
warning message with relevant information. At that point we pause
recovery, so a hot standby remains usable. We also repeat the last
warning message once a minute so it is harder to miss or ignore.
Reviewed-by: Sergei Kornilov <sk@zsrv.org>
Reviewed-by: Masahiko Sawada <masahiko.sawada@2ndquadrant.com>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/4ad69a4c-cc9b-0dfe-0352-8b1b0cd36c7b@2ndquadrant.com
The additional information displayed will be block number for error
occurring while processing heap and index name for error occurring
while processing the index.
This will help us in diagnosing the problems that occur during a vacuum.
For ex. due to corruption (either caused by bad hardware or by some bug)
if we get some error while vacuuming, it can help us identify the block
in heap and or additional index information.
It sets up an error context callback to display additional information
with the error. During different phases of vacuum (heap scan, heap
vacuum, index vacuum, index clean up, heap truncate), we update the error
context callback to display appropriate information. We can extend it to
a bit more granular level like adding the phases for FSM operations or for
prefetching the blocks while truncating. However, I felt that it requires
adding many more error callback function calls and can make the code a bit
complex, so left those for now.
Author: Justin Pryzby, with few changes by Amit Kapila
Reviewed-by: Alvaro Herrera, Amit Kapila, Andres Freund, Michael Paquier
and Sawada Masahiko
Discussion: https://www.postgresql.org/message-id/20191120210600.GC30362@telsasoft.com
Use IndexRelationGetNumberOfKeyAttributes() rather than
IndexRelationGetNumberOfAttributes() when determining whether or not two
index tuples are suitable for merging together into a single posting
list tuple. This is a little bit tidier. It brings affected code in
nbtdedup.c a little closer to similar, related code in nbtsplitloc.c.
Traditionally autovacuum has only ever invoked a worker based on the
estimated number of dead tuples in a table and for anti-wraparound
purposes. For the latter, with certain classes of tables such as
insert-only tables, anti-wraparound vacuums could be the first vacuum that
the table ever receives. This could often lead to autovacuum workers being
busy for extended periods of time due to having to potentially freeze
every page in the table. This could be particularly bad for very large
tables. New clusters, or recently pg_restored clusters could suffer even
more as many large tables may have the same relfrozenxid, which could
result in large numbers of tables requiring an anti-wraparound vacuum all
at once.
Here we aim to reduce the work required by anti-wraparound and aggressive
vacuums in general, by triggering autovacuum when the table has received
enough INSERTs. This is controlled by adding two new GUCs and reloptions;
autovacuum_vacuum_insert_threshold and
autovacuum_vacuum_insert_scale_factor. These work exactly the same as the
existing scale factor and threshold controls, only base themselves off the
number of inserts since the last vacuum, rather than the number of dead
tuples. New controls were added rather than reusing the existing
controls, to allow these new vacuums to be tuned independently and perhaps
even completely disabled altogether, which can be done by setting
autovacuum_vacuum_insert_threshold to -1.
We make no attempt to skip index cleanup operations on these vacuums as
they may trigger for an insert-mostly table which continually doesn't have
enough dead tuples to trigger an autovacuum for the purpose of removing
those dead tuples. If we were to skip cleaning the indexes in this case,
then it is possible for the index(es) to become bloated over time.
There are additional benefits to triggering autovacuums based on inserts,
as tables which never contain enough dead tuples to trigger an autovacuum
are now more likely to receive a vacuum, which can mark more of the table
as "allvisible" and encourage the query planner to make use of Index Only
Scans.
Currently, we still obey vacuum_freeze_min_age when triggering these new
autovacuums based on INSERTs. For large insert-only tables, it may be
beneficial to lower the table's autovacuum_freeze_min_age so that tuples
are eligible to be frozen sooner. Here we've opted not to zero that for
these types of vacuums, since the table may just be insert-mostly and we
may otherwise freeze tuples that are still destined to be updated or
removed in the near future.
There was some debate to what exactly the new scale factor and threshold
should default to. For now, these are set to 0.2 and 1000, respectively.
There may be some motivation to adjust these before the release.
Author: Laurenz Albe, Darafei Praliaskouski
Reviewed-by: Alvaro Herrera, Masahiko Sawada, Chris Travers, Andres Freund, Justin Pryzby
Discussion: https://postgr.es/m/CAC8Q8t%2Bj36G_bLF%3D%2B0iMo6jGNWnLnWb1tujXuJr-%2Bx8ZCCTqoQ%40mail.gmail.com
Delaying unlocking the right child page until after the point that the
left child's parent page has been refound is no longer truly necessary.
Commit 40dae7ec made nbtree tolerant of interrupted page splits. VACUUM
was taught to avoid deleting a page that happens to be the right half of
an incomplete split. As long as page splits don't unlock the left child
page until the end of the second/final phase, it should be safe to
unlock the right child page earlier (at the end of the first phase).
It probably isn't actually useful to release the right child's lock
earlier like this (it probably won't improve performance). Even still,
pointing out that it ought to be safe to do so should make it easier to
understand the overall design.
The parameters primary_conninfo, primary_slot_name and
wal_receiver_create_temp_slot can now be changed with a simple "reload"
signal, no longer requiring a server restart. This is achieved by
signalling the walreceiver process to terminate and having it start
again with the new values.
Thanks to Andres Freund, Kyotaro Horiguchi, Fujii Masao for discussion.
Author: Sergei Kornilov <sk@zsrv.org>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Álvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/19513901543181143@sas1-19a94364928d.qloud-c.yandex.net
Commit 3297308278 gave walreceiver the ability to create and use a
temporary replication slot, and made it controllable by a GUC (enabled
by default) that can be changed with SIGHUP. That's useful but has two
problems: one, it's possible to cause the origin server to fill its disk
if the slot doesn't advance in time; and also there's a disconnect
between state passed down via the startup process and GUCs that
walreceiver reads directly.
We handle the first problem by setting the option to disabled by
default. If the user enables it, its on their head to make sure that
disk doesn't fill up.
We handle the second problem by passing the flag via startup rather than
having walreceiver acquire it directly, and making it PGC_POSTMASTER
(which ensures a walreceiver always has the fresh value). A future
commit can relax this (to PGC_SIGHUP again) by having the startup
process signal walreceiver to shutdown whenever the value changes.
Author: Sergei Kornilov <sk@zsrv.org>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Álvaro Herrera <alvherre@alvh.no-ip.org>
Discussion: https://postgr.es/m/20200122055510.GH174860@paquier.xyz
Previously if a promotion was triggered while recovery was paused,
the paused state continued. Also recovery could be paused by executing
pg_wal_replay_pause() even while a promotion was ongoing. That is,
recovery pause had higher priority over a standby promotion.
But this behavior was not desirable because most users basically wanted
the recovery to complete as soon as possible and the server to become
the master when they requested a promotion.
This commit changes recovery so that it prefers a promotion over
recovery pause. That is, if a promotion is triggered while recovery
is paused, the paused state ends and a promotion continues. Also
this commit makes recovery pause functions like pg_wal_replay_pause()
throw an error if they are executed while a promotion is ongoing.
Internally, this commit adds new internal function PromoteIsTriggered()
that returns true if a promotion is triggered. Since the name of
this function and the existing function IsPromoteTriggered() are
confusingly similar, the commit changes the name of IsPromoteTriggered()
to IsPromoteSignaled, as more appropriate name.
Author: Fujii Masao
Reviewed-by: Atsushi Torikoshi, Sergei Kornilov
Discussion: https://postgr.es/m/00c194b2-dbbb-2e8a-5b39-13f14048ef0a@oss.nttdata.com
restore_command has only been used until now by the backend, but there
is a pending patch for pg_rewind to make use of that in the frontend.
Author: Alexey Kondratov
Reviewed-by: Andrey Borodin, Andres Freund, Alvaro Herrera, Alexander
Korotkov, Michael Paquier
Discussion: https://postgr.es/m/a3acff50-5a0d-9a2c-b3b2-ee36168955c1@postgrespro.ru
This commit introduces new wait events BackupWaitWalArchive and
RecoveryPause. The former is reported while waiting for the WAL files
required for the backup to be successfully archived. The latter is
reported while waiting for recovery in pause state to be resumed.
Author: Fujii Masao
Reviewed-by: Michael Paquier, Atsushi Torikoshi, Robert Haas
Discussion: https://postgr.es/m/f0651f8c-9c96-9f29-0ff9-80414a15308a@oss.nttdata.com
Until now, only selected bulk operations (e.g. COPY) did this. If a
given relfilenode received both a WAL-skipping COPY and a WAL-logged
operation (e.g. INSERT), recovery could lose tuples from the COPY. See
src/backend/access/transam/README section "Skipping WAL for New
RelFileNode" for the new coding rules. Maintainers of table access
methods should examine that section.
To maintain data durability, just before commit, we choose between an
fsync of the relfilenode and copying its contents to WAL. A new GUC,
wal_skip_threshold, guides that choice. If this change slows a workload
that creates small, permanent relfilenodes under wal_level=minimal, try
adjusting wal_skip_threshold. Users setting a timeout on COMMIT may
need to adjust that timeout, and log_min_duration_statement analysis
will reflect time consumption moving to COMMIT from commands like COPY.
Internally, this requires a reliable determination of whether
RollbackAndReleaseCurrentSubTransaction() would unlink a relation's
current relfilenode. Introduce rd_firstRelfilenodeSubid. Amend the
specification of rd_createSubid such that the field is zero when a new
rel has an old rd_node. Make relcache.c retain entries for certain
dropped relations until end of transaction.
Back-patch to 9.5 (all supported versions). This introduces a new WAL
record type, XLOG_GIST_ASSIGN_LSN, without bumping XLOG_PAGE_MAGIC. As
always, update standby systems before master systems. This changes
sizeof(RelationData) and sizeof(IndexStmt), breaking binary
compatibility for affected extensions. (The most recent commit to
affect the same class of extensions was
089e4d405d0f3b94c74a2c6a54357a84a681754b.)
Kyotaro Horiguchi, reviewed (in earlier, similar versions) by Robert
Haas. Heikki Linnakangas and Michael Paquier implemented earlier
designs that materially clarified the problem. Reviewed, in earlier
designs, by Andrew Dunstan, Andres Freund, Alvaro Herrera, Tom Lane,
Fujii Masao, and Simon Riggs. Reported by Martijn van Oosterhout.
Discussion: https://postgr.es/m/20150702220524.GA9392@svana.org
The function assumed forkNum=MAIN_FORKNUM and page_std=true, ignoring
the actual arguments. Existing callers passed exactly those values, so
there's no live bug. Back-patch to v12, where the function first
appeared, because another fix needs this.
Discussion: https://postgr.es/m/20191118045434.GA1173436@rfd.leadboat.com
swap_relation_files() calls toast_get_valid_index() to find and lock
this index, just before swapping with the rebuilt TOAST index. The
latter function releases the lock before returning. Potential for
mischief is low; a concurrent session can issue ALTER INDEX ... SET
(fillfactor = ...), which is not alarming. Nonetheless, changing
pg_class.relfilenode without a lock is unconventional. Back-patch to
9.5 (all supported versions), because another fix needs this.
Discussion: https://postgr.es/m/20191226001521.GA1772687@rfd.leadboat.com
Remove comments that are a throw back to a time when nbtree cared about
write-ordering dependencies. The comments are similar to those removed
by commit 9ee7414e, among others.
This commit renames RecoveryWalAll and RecoveryWalStream wait events to
RecoveryWalStream and RecoveryRetrieveRetryInterval, respectively,
in order to make the names and what they are more consistent. For example,
previously RecoveryWalAll was reported as a wait event while the recovery
was waiting for WAL from a stream, and which was confusing because the name
was very different from the situation where the wait actually could happen.
The names of macro variables for those wait events also are renamed
accordingly.
This commit also changes the category of RecoveryRetrieveRetryInterval to
Timeout from Activity because the wait event is reported while waiting based
on wal_retrieve_retry_interval.
Author: Fujii Masao
Reviewed-by: Kyotaro Horiguchi, Atsushi Torikoshi
Discussion: https://postgr.es/m/124997ee-096a-5d09-d8da-2c7a57d0816e@oss.nttdata.com
Only internal page splits need to call _bt_pgaddtup() instead of
PageAddItem(), and only for data items, one of which will end up at the
first offset (or first offset after the high key offset) on the new
right page. This data item alone will need to be truncated in
_bt_pgaddtup().
Since there is no reason why retail inserts ever need to truncate the
incoming item, use a raw PageAddItem() call there instead. Even
_bt_split() uses raw PageAddItem() calls for left page and right page
high keys. Clearly the _bt_pgaddtup() shim function wasn't really
encapsulating anything. _bt_pgaddtup() should now be thought of as a
_bt_split() helper function.
Note that the assertions from commit d1e241c2 verify that retail inserts
never insert an item at an internal page's negative infinity offset.
This invariant could only ever be violated as a result of a basic logic
error in nbtinsert.c.
Commit 2b272734, which added the fastpath rightmost leaf page cache
insert optimization, added code to _bt_doinsert() to handle using and
invalidating the backend local block cache. It doesn't seem like a good
place to handle these low level details, though. _bt_doinsert() is
supposed to be a high level function -- it is the main entry point to
nbtinsert.c.
Restructure the code by placing handling of the rightmost block cache at
the start of a new _bt_search() shim function, _bt_search_insert(). The
new function is called from _bt_doinsert(), which uses it as a
_bt_search() variant that conveniently accepts its BTInsertState state
as an argument. _bt_doinsert() no longer needs to directly consider the
fastpath optimization.
Discussion: https://postgr.es/m/CAH2-Wzk59cxKJRd=rfbyub6-V4yWRjsOYRkUNHBLT1P1GdtCQQ@mail.gmail.com
Copying block and offset numbers to local variables in _bt_insertonpg()
made the code less readable. Remove the variables. There is already
code that conditionally calls BufferGetBlockNumber() in the same block,
so consistently do it that way instead.
Calling BufferGetBlockNumber() is very cheap, but we might as well avoid
it when it isn't truly necessary. It isn't truly necessary for
_bt_insertonpg() to call BufferGetBlockNumber() in almost all cases.
Spotted while working on a patch that refactors the fastpath rightmost
leaf page cache optimization, which was added by commit 2b272734.
We don't need to manually clean up allocations in a SRF's
multi_call_memory_ctx, because the SRF_RETURN_DONE infrastructure
takes care of that (and also ensures that it will happen even if the
function never gets a final call, which simple manual cleanup cannot
do).
Hence, the code removed by this patch is a waste of code and cycles.
Worse, it gives the impression that cleaning up manually is a thing,
which can lead to more serious errors such as those fixed in
commits 085b6b667 and b4570d33a. So we should get rid of it.
These are not quite actual bugs though, so I couldn't muster the
enthusiasm to back-patch. Fix in HEAD only.
Justin Pryzby
Discussion: https://postgr.es/m/20200308173103.GC1357@telsasoft.com
Refactor nbtinsert.c so that the final itemsz of each new non-pivot
tuple (the MAXALIGN()'d size) is determined once. Most of the functions
used by leaf page inserts used the insertstate.itemsz value already.
This commit makes everything use insertstate.itemsz as standard
practice. The goal is to decouple tuple size from "effective" tuple
size. Making this distinction isn't truly necessary right now, but that
might change in the future.
Also explain why we consistently apply MAXALIGN() to get an effective
index tuple size. This was rather unclear, in part because it isn't
actually strictly necessary right now.
Peter Geoghegan
Tom Lane
Alvaro Herrera
Michael Paquier
Fujii Masao
Amit Kapila
Peter Eisentraut
Thomas Munro
Alexander Korotkov
David Rowley
Andres Freund
Tomas Vondra
Noah Misch
Robert Haas